While I was initially impressed with it's context window, I got so sick of fighting with Claude about what it was allowed to answer I quit my subscription after 3 months.
Their whole policing AI models stance is commendable but ultimately renders their tools useless.
It actually started arguing with me about whether it was allowed to help implement a github repository's code as it might be copywritten... it was MIT licensed open source from Google :/
I just include text that I own the device in question and that I have a legal team watching my every move. It's stupid, I agree, but not insurmountable. I had less refusals with Claude 3 Opus.
Their whole policing AI models stance is commendable but ultimately renders their tools useless.
It actually started arguing with me about whether it was allowed to help implement a github repository's code as it might be copywritten... it was MIT licensed open source from Google :/